1. Field of the Invention
This invention relates generally to structures, essentially in one-dimensional form, and which are of nanometer dimensions in their width or diameter, and which are commonly known as nanowhiskers, nanorods, nanowires, nanotubes, etc.; for the purposes of this specification, such structures will be termed “one-dimensional nanoelements”. More specifically, but not exclusively, the invention relates to nanowhiskers, and to methods of forming nanowhiskers.
2. Brief Description of the Prior Art
The basic process of whisker formation on substrates, by the so-called VLS (vapor-liquid-solid) mechanism is well known. A particle of a catalytic material, usually gold, for example, on a substrate is heated in the presence of certain gases to form a melt. A pillar forms under the melt, and the melt rises up on top of the pillar. The result is a whisker of a desired material with the solidised particle melt positioned on top—see “Growth of Whiskers by the Vapour-Liquid-Solid Mechanism”—Givargizov—Current Topics in Materials Science Vol. 1, pages 79-145—North Holland Publishing Company 1978. The dimensions of such whiskers were in the micrometer range.
International Application WO 01/84238 discloses in FIGS. 15 and 16 a method of forming nanowhiskers wherein nanometer sized particles from an aerosol are deposited on a substrate and these particles are used as seeds to create filaments or nanowhiskers. For the purposes of this specification the term nanowhiskers is intended to mean one dimensional nanoelements with a diameter of nanometer dimensions, the element having been formed by the VLS mechanism.
Typically, nanostructures are devices having at least two dimensions less than about 1 μm (i.e., nanometer dimensions). Ordinarily, layered structures or stock materials having one or more layers with a thickness less than 1 μm are not considered to be nanostructures, although nanostructures may be used in the preparation of such layers, as is disclosed below. Thus the term nanostructures includes free-standing or isolated structures having two dimensions less than about 1 μm which have functions and utilities that are different from larger structures and are typically manufactured by methods that are different from conventional procedures for preparing somewhat larger, i.e., microscale, structures. Thus, although the exact boundaries of the class of nanostructures are not defined by a particular numerical size limit, the term has come to signify such a class that is readily recognized by those skilled in the art. In many cases, an upper limit of the size of the dimensions that characterize nanostructures is about 500 nm.
Where the diameter of a nanoelement is below a certain amount, say 50 nm, quantum confinement occurs where electrons can only move in the length direction of the nanoelement; whereas for the diametral plane, the electrons occupy quantum mechanical eigenstates.
The electrical and optical properties of semiconductor nanowhiskers are fundamentally determined by their crystalline structure, shape, and size. In particular, a small variation of the width of the whisker may provoke a considerable change in the separation of the energy states due to the quantum confinement effect. Accordingly, it is of importance that the whisker width can be chosen freely, and, of equal importance, is that the width can be kept constant for extended whisker lengths. This, together with the possibility of positioning whiskers at selected positions on a substrate, will be necessary if an integration of whisker technology with current semiconductor component technology is to be possible. Several experimental studies on the growth of GaAs whiskers have been made, the most important reported by Hiruma et al. They grew III-V nano-whiskers on III-V substrates in a metal organic chemical vapor deposition -MOCVD- growth system—K. Hiruma, M. Yazawa, K. Haraguchi, K. Ogawa, T. Katsuyama, M. Koguchi, and H. Kakibayashi, J. Appl. Phys. 74, 3162 1993; K. Hiruma, M. Yazawa, T. Katsuyama, K. Ogawa, K. Haraguchi, M. Koguchi, and H. Kakibayashi, J. Appl. Phys. 77, 447 1995; E. I. Givargizov, J. Cryst. Growth 31, 20 1975; X. F. Duan, J. F. Wang, and C. M. Lieber, Appl. Phys. Lett. 76, 1116 2000; K. Hiruma, H. Murakoshi, M. Yazawa, K. Ogawa, S. Fukuhara, M. Shirai, and T. Katsuyama, IEICE Trans. Electron. E77C, 1420 1994; K. Hiruma, et al, “Self-organised growth on GaAs/InAs heterostructure nanocylinders by organometallic vapor phase epitaxy”, J. Crystal growth 163, (1996), 226-231. Their approach relied on annealing a thin Au film to form the seed particles. In this way, they achieved a homogeneous whisker width distribution, the mean size of which could be controlled by the thickness of the Au layer and the way this layer transforms to nanoparticles. With this technique, it is difficult to control the size and surface coverage separately, and it is virtually impossible to achieve a low coverage. The correlation between film thickness and whisker thickness was not straightforward, since the whisker width also depended on growth temperature, and there were even signs of a temperature-dependent equilibrium size of the Au particles. The authors also noticed a strong correlation between the size of the Au droplets de-posited from a scanning tunneling microscope tip and the resulting whisker width. For the free-flying Si whiskers grown by Lieber et al.,—Y. Cui, L. J. Lauhon, M. S. Gudiksen, J. F. Wang, and C. M. Lieber, Appl. Phys. Lett. 78, 2214, 2001—a clear particle-whisker size correlation has been shown.
It is necessary, if whiskers are to be used as electrical components, that there should be well-defined electrical junctions situated along the length of a whisker, and much work has been directed at achieving this—see for example Hiruma et al, “Growth and Characterisation of Nanometer-Scale GaAs, AlGaAs and GaAs/InAs Wires” IEICE Trans. Electron., Vol. E77-C, No. 9 September 1994, pp 1420-1424. However, much improvement is necessary.
Much work has also been carried out on carbon nanotubes (CNTs). Despite progress, research has been frustrated by a lack of control of the conductivity-type of CNTs and an inability to form 1D heterostructures in a controlled manner. Randomly formed interfaces as kinks between metallic and semiconducting parts of CNTs have been identified and studied (Yao et al, Nature, 1999, 402, 273) as have doping (pn) junctions in semiconducting CNTs (Derycke et al, Nano Letters, 2001, 1, 453) and transitions between CNTs and semiconductor (Si and SiC) nanowhiskers (Hu et al, Nature, 1999, 399, 48).
In a separate trend of development, attempts to fabricate 1D devices have been made since the late 1980s by top-down methods, as pioneered by Randall, Reed and co-workers at Texas Instruments—M. A. Reed et al., Phys. Rev. Lett. 60, 535(1988). Their top-down approach, which still represents the state of the art for this family of quantum devices, is based on epitaxial growth of multi-layers defining the two barriers and the central quantum well. Electron-beam lithography is then used to define the lateral confinement pattern, together with evaporation of the metallic layers to form the top contact. A lift-off process is then used to remove the e-beam-sensitive resist from the surface, and reactive ion etching removes all the material surrounding the intended narrow columns. Finally, the devices are contacted via the substrate and from the top using a polyimide layer. In the studies of devices fabricated by this bottom-up technique, 100-200 nm diameter columns have been observed, however, with rather disappointing electrical characteristics and peak-to-valley currents at best around 1.1:1. An alternative approach to realizing low-dimensional resonant tunneling devices has been reported more recently, employed strain-induced formation of self-assembled quantum dots (I. E. Itskevich et al., Phys. Rev. B 54, 16401(1996); M. Narihiro, G. Yusa, Y. Nakamura, T. Noda, H. Sakaki, Appl. Phys. Lett. 70, 105(1996); M. Borgstrom et al., Appl. Phys. Lett. 78, 3232(2001)).